CERT-In flags AI-driven cyberattacks on payment systems beyond banks
Synopsis
Key Takeaways
The Indian Computer Emergency Response Team (CERT-In) has issued a formal advisory warning that AI-powered cyberattacks on payment systems have spread well beyond banks and financial institutions, now targeting technology service providers that support government-linked payment processes and platforms managing vendor or partner payouts. The alert, dated 21 November, marks a significant escalation in the threat landscape for India's digital payment infrastructure.
How the Attacks Work
Unlike conventional cyberattacks that directly target payment application programming interfaces (APIs), the new wave of intrusions focuses on manipulating business records and approval workflows that determine whether payments are authorised. By altering these upstream processes, cybercriminals can divert funds through what appear to be entirely legitimate payment channels — making detection considerably harder.
According to CERT-In's advisory, the concern is no longer confined to API security alone. 'A transaction may pass the usual application and approval checks even when the records used to initiate it have been changed without authorisation,' the advisory stated. Attackers are exploiting widely used programming languages, particularly Java and PHP, which together underpin roughly 70 per cent of the internet.
The AI-Driven Offensive Framework
CERT-In has specifically flagged the deployment of what it terms an 'AI-driven agentic offensive framework' (AOF) — a coordinated system of multiple automated agents or 'workers' that simultaneously carry out reconnaissance, privilege discovery, and other intrusion tasks in parallel. This approach dramatically reduces the manual effort required by attackers and compresses the time between initial reconnaissance and a full-scale breach.
'AOFs allow several intrusion tasks to be carried out with less manual effort and can shorten the time between reconnaissance and subsequent attack stages,' the advisory noted. CERT-In has urged all organisations — irrespective of sector — to reassess their exposure, access controls, and monitoring arrangements in light of this threat.
Who Is at Risk
The threat extends across a broad range of entities. CERT-In has identified enterprises, technology service providers, government organisations, and academic and research institutions as active targets. Earlier in November, the agency had separately warned that hackers were using AI tools to breach live systems of Indian non-banking financial corporations (NBFCs), fintech companies, and payment aggregators to steal funds directly.
Notably, the shift from targeting financial sector firms to government-linked payment platforms represents a qualitative change in attacker ambition. Because these transactions flow through trusted systems and established procedures, suspicious activity is difficult to detect in real time.
What Organisations Should Do
CERT-In's advisory calls on organisations to conduct thorough reviews of their business payment workflows, not just their API security layers. Access controls, audit trails for record modifications, and anomaly-detection monitoring are highlighted as priority areas. The agency's guidance is directed at all sectors, signalling that no industry vertical should consider itself outside the threat perimeter.
As AI tooling becomes more accessible to threat actors, security experts warn that the velocity and sophistication of such attacks are likely to increase — placing sustained pressure on India's digital payment ecosystem in the months ahead.